py-channels: updated to 4.3.2
4.3.2 (2025-11-20)
* Confirmed support for Django 6.0.
* Confirmed support for Python 3.14.
* Added ``types`` extra for ``types-channels`` stubs. See installation docs.
The hp300 reserves the last page of RAM to use for the MMU trampoline.
Since this page would otherwise be wasted, put it to good use as the
NULL segment table using a new NULL_SEGTAB_PA hook.
devel/task: update to 3.45.5
* Fixed bug that made a generic message, instead of an useful one, appear
when a Taskfile could not be found.
* Fixed a bug that caused an error when including a Remote Git Taskfile.
* Fixed issue where .taskrc.yml was not returned if reading it failed, and
corrected handling of remote entrypoint Taskfiles.
* Improved performance of --list and --list-all by introducing a faster
compilation method that skips source globbing and checksum updates.
* Fixed a concurrency bug with output: group. This ensures that begin/end
parts won't be mixed up from different tasks.
* Do not re-evaluate variables for defer:.
* Improve error message when a Taskfile is not found.
* Fixed generic error message exit status 1 when a dependency task failed.
* Fixed YAML library from the unmaintained gopkg.in/yaml.v3 to the new fork
maintained by the official YAML org.
* On Windows, the built-in version of the rm core utils contains a fix
related to the -f flag.
The Hibler/Utah pmap put the virtual kernel PTE array near the top of
the address space because it needed to use the very last page of kernel
virtual space to map the last page of RAM VA==PA to faciliate MMU on/off
transitions, and doing so made for more efficient use of kernel PT pages.
We use the SYSMAP_VA hook to do the same for the same reason.
(The *2 is because we can't use the very top of the address space
for the virtual kernel PTE array because MAXADDR is otherwise occupied.)
q-dns: update to 0.19.11
* Add Windows zip format override in goreleaser config
* Fix format for Windows archive in goreleaser config
* Fix formatting for Windows archive in goreleaser config
* Merge pull request #140 from FaffeF/FaffeF-patch-archive-format
* ci: update binary usage in README
* ci: update coverage
* fix: connection error on different platforms
* fix: error on multiple servers (#92)
Add a FIXEDVA flag to the pmap_bootmap structure that can be used to
map a physical address at a fixed virtual location.
This is intended to support the somewhat quirky physical/virtual address
space layout of the hp300 where the last page of physical memory is
mapped VA==PA at the very top of the virtual address space in order to
facilitate enabling / disabling the MMU.
py-pre-commit: update to 4.5.0
4.5.0 - 2025-11-22
Features
* Add pre-commit hazmat.
4.4.0 - 2025-11-08
Features
* Add --fail-fast option to pre-commit run.
* Upgrade ruby-build / rbenv.
* Add language: unsupported / language: unsupported_script as aliases for
language: system / language: script (which will eventually be deprecated).
* Add support docker-in-docker detection for cgroups v2.
Fixes
[3 lines not shown]
iosevka-*: update to 33.3.5
Changes of version 33.3.5
* Refine shape of the following characters:
* CYRILLIC CAPITAL LETTER DJE (U+0402).
* CYRILLIC CAPITAL LETTER TSHE (U+040B).
* CYRILLIC CAPITAL LETTER YAT (U+0462).
* CYRILLIC SMALL LETTER YAT (U+0463).
* CYRILLIC CAPITAL LETTER SHHA (U+04BA).
* CYRILLIC CAPITAL LETTER SHHA WITH DESCENDER (U+0526).
* CYRILLIC SMALL LETTER TALL YAT (U+1C87).
* COMBINING CYRILLIC LETTER YAT (U+2DFA).
* CYRILLIC CAPITAL LETTER IOTIFIED YAT (U+A652).
* CYRILLIC SMALL LETTER IOTIFIED YAT (U+A653).
* CYRILLIC CAPITAL LETTER HWE (U+A694).
* Refine spacing of diagonal-tailed i and j.
* Add Characters:
* LATIN CAPITAL LETTER VY (U+A760).
[19 lines not shown]
Work-in-progress. It's not quite working yet, but if the option isn't
defined, everyhing still works:
In the Hibler/Utah pmap, the kernel PTE array was placed right near
the very top of the kernel virtual address space. This was because
of the hp300's unique physical memory arrangement: the last page of
memory is always located at PA $FFFF.F000 and the physical address
of the beginning of RAM varied based on the RAM size. This meant that
VA $FFFF.F000 is a convenient place to map the RAM VA==PA, making
transition between "MMU off" and "MMU on" (and vice versa) easier.
Since VA $FFFF.F000 was already going to be mapped, it made sense to
put something else along side of it in order to minimize waste in
PT pages.
As noted above, this implementation is tuned for a growing-from-0
virtual space layout. However, we have a special case for this
particular requirement: if a platform defines SYSMAP_VA, then we
will assume it is as a high address, place the kernel PTE array at
that KVA, and ensure sufficient page tables to map from that VA until
[20 lines not shown]
lxqt-build-tools: expect C++20 now, we really need it
LXQt includes a (as we call it) plasma6-libkscreen dependency, which as
of its version 6.5.2 or thereabouts expects C++20 features (the
"spaceship operator", at the least). Fixes builds of lxqt-config after
the latest plasma6 updates earlier today.
aes(9): Rewrite x86 SSE2 implementation.
This computes eight AES_k instances simultaneously, using the
bitsliced 32-bit aes_ct logic which computes two blocks at a time in
uint32_t arithmetic, vectorized four ways.
Previously, the SSE2 code was a very naive adaptation of aes_ct64,
which computes four blocks at a time in uint64_t arithmetic, without
any 2x vectorization -- I did it at the time because:
(a) it was easier to get working,
(b) it only affects really old hardware with neither AES-NI nor SSSE3
which are both much much faster.
But it was bugging me that this was a kind of dumb use of SSE2.
Substantially reduces stack usage (from ~1200 bytes to ~800 bytes)
and should approximately double throughput for CBC decryption and for
XTS encryption/decryption.
[10 lines not shown]
aes(9): New 64-bit bitsliced implementation.
Derived from BearSSL's aes_ct64 code. Compared to the aes_ct code,
on machines with native 64-bit integer arithmetic, aes_ct64 should
have approximately:
- the same throughput for:
. CBC encryption,
. CCM encryption/decryption, and
. CBC-MAC;
- double the throughput for:
. CBC decryption,
. XTS encryption/decryption.
(aes_ct computes AES on two blocks at a time; aes_ct64 computes it on
four blocks at a time, with roughly the same number of instructions.
CBC encryption and CBC-MAC are inherently sequential; CCM, being a
combination of CTR and CBC-MAC, can only really be parallelized two
[12 lines not shown]
libpthread: Link with -Wl,-z,nodelete.
Can't safely unload libpthread because of the interaction with libc
thread stubs.
PR lib/59784: dlopening and dlclosing libpthread is broken
